Article

TaskWAM uses stage-aware tokens to steer world-action models toward intended futures

The paper proposes TaskWAM, a closed-loop, stage-aware framework that uses Mode Tokens to represent fine-grained task requirements and State Tokens to track execution progress, forming a structured task state that steers world-action model generation toward futures matching the current task, addressing the Stage Confusion phenomenon in long-horizon manipulation.

Part of the event: Proposes TaskWAM framework and StageBench benchmark

This page provides a summary and source link only.